Health insurance continuity and survival among children, adolescents, and young adults with blood cancer: An analysis based on SEER-Medicaid data.

Abstract

6504 Background: Many previously uninsured patients do not receive Medicaid coverage until the point of, or after, cancer diagnosis, which may delay access to life-saving therapies. We examined whether continuous Medicaid coverage prior to diagnosis (vs. gaining Medicaid at, or after, diagnosis) provides a survival benefit in children and adolescents/young adults (AYAs) newly diagnosed with blood cancer. Methods: Using the linked SEER cancer registry-Medicaid enrollment data, we assessed coverage during the six months prior to, the month of, and the six months after cancer diagnosis (13-month window) in 29,873 children and AYAs (aged 0-39 years) diagnosed with blood cancer in 2006-2013. To measure insurance continuity, we categorized patients who had: (1) continuous Medicaid (enrolled for ≥6 months prior to through diagnosis), (2) newly gained Medicaid at/after diagnosis (enrollment only at or ≤6 months after diagnosis), (3) other patterns of noncontinuous Medicaid, (4) private insurance, (5) other insurance, or (6) uninsured/unknown insurance. The last three groups contain patients in SEER not linked to Medicaid enrollment data; these patients’ insurance at diagnosis was available in SEER. Kaplan-Meier (KM) survival curve of 5-year overall survival was stratified by insurance continuity category. Associations between insurance continuity and survival were examined using Cox proportional hazard models, for all blood cancers combined and by cancer type (leukemia, lymphoma, myeloma). Models also adjusted for age, sex, race/ethnicity, rurality, neighborhood socioeconomic status, and diagnosis year. Results: Of our sample, 44.1% had private insurance, followed by Medicaid (28.3%), other insurance (9.6%), and no insurance (3.9%) or unknown insurance (14.1%). Of Medicaid-insured patients, 44.9% had continuous Medicaid, 38.9% gained Medicaid at/after diagnosis, and 16.2% experienced other noncontinuous patterns. The KM-estimated 5-year survival probability was 89.3% for privately insured patients, followed by other insurance (88.1%), continuous Medicaid (78.2%), other patterns of noncontinuous Medicaid (74.3%), and newly gained Medicaid at/after diagnosis (70.6%; p<0.001). In multivariable analysis, patients who newly gained Medicaid at/after diagnosis had a hazard ratio (HR) for death of 2.63 (95%CI=2.42-2.87), whereas patients with continuous Medicaid and those with other noncontinuous patterns had HRs of 2.08 (95%CI=1.90-2.28) and 2.31 (95%CI=2.04-2.60), respectively, compared to privately insured patients. This finding persisted across blood cancer types. Conclusions: Less than half of Medicaid-insured children and AYAs with blood cancer had continuous insurance coverage prior to and following cancer diagnosis. Lacking continuous Medicaid coverage was associated with inferior survival.